The Continuum of Psychosis

The psychotic disorders include schizophrenia, bipolar affective disorder, schizophreniform disorder, delusional disorder; drug induced psychotic disorder, brief psychotic disorder and psychotic disorder not otherwise specified. Psychotic disorders are characterised by psychotic symptoms (hallucinations and delusions), motivational impairment, changes in affect (blunted or inappropriate affect) and impairments in cognitive functioning (1). Psychosis sits on a continuum spanning from those who have no hallucinations or delusions through to people with psychotic disorders. In between these two ends of the psychosis continuum are those with hallucinations and other sub-threshold psychotic symptoms commonly known as psychotic-like experiences (PLE) (2).

We reported from the first ever national survey of hallucinations in adolescents that almost 10% of Australian adolescents experienced hallucinations. Hallucinations were associated with living in a non-intact family, depression, internalising and externalising symptoms and cannabis use (3)*. In other studies of Australian adults, we found that more than 1 in 12 Australian adults reported PLE. Risk factors associated with PLE were very similar to those for the full clinical syndrome of schizophrenia (4, 5)*. Those who experienced PLE were more likely to suffer from depression, anxiety, less social supports and socio-economic disadvantage. They had a four-fold increased risk of suicidal ideation and attempts (6-9)*. Based on the MUSP cohort, we identified, unexpectedly, steep age differences with respect to PLE in young adulthood; those aged 18-20 years were significantly more likely to experience delusions compared to those aged 21-23 years (11.7% vs 8.0%) (10)*. We speculate that there may be a critical transition period during adolescence and young adulthood with respect to the expression of psychosis risk, perhaps as a reflection of underlying brain maturational processes. This proposed study will examine the adult outcomes of participants who have hallucinations in his critical transitional period.
There has been a great deal of interest and contention surrounding the notion of a continuum of psychosis (2). A DSM-V diagnostic category of ‘risk syndrome for psychosis (encapsulating those on the psychosis continuum) has been proposed (11); however, high rates of false positive outcomes and unnecessary exposure to stigma, discrimination, anxiety and antipsychotic medication has made many cautious of this new diagnostic category (12). World class Australian research has shown that even in help seeking adolescents and young adults who meet criteria of high risk for psychosis (e.g. those experiencing brief limited psychotic symptoms, attenuated psychotic symptoms or have lowered mood/ anxiety combined with schizotypal personality or family history) only a minority (20-40%) will go on to develop a psychotic disorder in the subsequent two years (13, 14). Longitudinal, population-based studies of the course of hallucinations across adolescence and young adulthood are necessary in order to understand the factors that lead some individuals to develop a psychotic disorder, others to experience depression and anxiety and some to be free of mental ill health.

The Clinical Trajectory of Hallucinations

The identification of individuals at risk of psychosis has been an area of intense research with Australian researchers leading the world (13, 15). With the aim of preventing later disability, the clinical outcomes of PLE in children and adolescents are of particular interest (16). Self-reported hallucinations in community populations have been shown to be the most reliable and strongest predictors of actual clinician assessed psychotic symptoms (17). To date, only three studies have examined the longitudinal outcomes of hallucinations in children and adolescents into adulthood. Two studies found hallucinations in adolescents were associated with increased risk of psychosis in adulthood. Participants in a New Zealand cohort with PLE in childhood were at increased risk of adult schizophreniform disorder and anxiety disorders but not depression or substance abuse (18). Our research from MUSP (n=3801) found that hallucinations at 14 years significantly increased the risk of non-affective psychosis at age 21 (Odds ratio and 95% CI: Males 5.09; 2.18, 11.84; Females 2.27; 1.01, 5.12) (19)*(20)* however an important limitation of this study was that participants had not passed through the high risk period for psychosis.
In contrast, a Dutch study (21) found no association between hallucinations in adolescence and psychotic illness 8 years later. Adolescents with hallucinations did have approximately twice the risk of developing depressive disorders and substance use disorders. However, this study was underpowered (n=783) to detect low prevalence disorders such as psychosis and many of the participants had not passed through the period of greatest risk of developing a psychotic disorder.
No studies to date have looked at other outcomes of adolescents and young adults who experience hallucinations. We do not know if these young people with hallucinations will later experience social and occupational disadvantage and increased risk of suicidality. The proposed study will investigate broad outcomes at 30 years of participants who reported experiencing hallucinations (a) at 14 years and (b) at 21 year and (c) persistent hallucinations. MUSP is one of the few datasets in the world with a cohort that is large enough to examine both common and low prevalence (i.e. schizophrenia) outcomes of hallucinations. We will also have the capacity to explore outcomes of common non-psychotic mental disorders (anxiety, depression, substance abuse and personality disorders). This will address an important gap in our understanding of the clinical significance of hallucinations in adolescents and young adults. However, one of the challenges of conducting such a study is to use clinically valid instruments to assess psychotic symptoms and disorders.

Assessing Psychotic Symptoms and Disorders

Most large epidemiological surveys have assessed psychotic symptoms and disorders using the Composite International Diagnostic Interview (CIDI) (2, 5, 22) a diagnostic instrument administered by lay interviewers. The earliest studies using the CIDI 1.0 on large community populations revealed high rates of false positive endorsement of psychosis items. For example, in the first United States National Comorbidity Survey, 28.4% of the participants endorsed one or more psychotic symptoms (23).
To overcome this problem, more recent versions of the CIDI was reworded so as to improve the validity of the psychosis items. Using the CIDI 3.0 in the US National Comorbidity Replication Study, auditory hallucinations and visual hallucinations were endorsed by 4.0% and 6.3% of the 2322 participants surveyed respectively. Clinicians recoded the subject’s text responses. The Positive Predictive Value (PPV) of CIDI endorsed psychotic items to clinician recoded psychotic symptoms was 38.5% (22), indicating that there will still be false positives identified using this CIDI. Both Kendler and Kessler have recommended that rather than relying on th CIDI, clinical interviews are necessary to reliably assess for psychotic symptoms and psychotic disorders (24), (22).
In summary, research to date has shown that hallucinations are common in adolescence and young adults, reflecting a continuum of psychosis. Clinically, there is a great deal of interest in the trajectory of young people with hallucinations as a better understanding of the outcomes may inform opportunities for prevention and early intervention. The few available longitudinal studies of adolescents with hallucinations have only examined narrow outcomes such as psychotic disorders and have not investigated outcomes of broader psychosocial functioning. The studies have been hampered by the use of lay interviews (i.e. CIDI) rather than clinician measures. Those young people with persistent hallucinations purported to be at particularly high risk of adverse outcomes however, to date there are a paucity of studies to further inform this hypothesis.


  1. To investigate whether hallucinations at 14 years and/or 21 years of age increases the risks for both psychotic and non-psychotic disorders at 30 years of age.
  2. To investigate at 30 years, the risk of suicidality; impairment in social and occupational functioning and reduced quality of life in those who experienced hallucinations at 14 or 21 years.
  3. To investigate if there is an increased risk of adverse mental health and psycho-social outcomes in those participants experienced hallucinations at both 14 and 21 years (persistent hallucinations) compared with those participants who only experienced hallucinations at one time point and those who never experienced hallucinations.

The new study began in January 2013.